임의 지역의 태양광발전소 일사량 추정을 위한 공간통계 보간 기술의 활용 및 사례 연구

이예지,최두성,고명진 대한건축학회 2022 대한건축학회논문집 Vol.38 No.9

In order to predict and manage the amount of power output of a photovoltaic system in a distributed grid, solar radiation prediction is essential. In order to increase the accuracy of the solar radiation prediction model, data measured at the target location where the power plant is located should be used. However, if there is no observation data, public data such as ASOS and AWS operated by the government can be effectively utilized. if the target location is far away from the station, uncertainty in the prediction is expected to increase due to the difference in distance. In this study, in order to solve this problem, solar radiation was estimated using inverse distance weighted interpolation (IDW), a spatial statistical technique that can estimate the values of unsampled locations. In addition, the possibility of application of the inverse distance weighted interpolation (IDW) was confirmed by validating the prediction model as a case study of six solar power plants in operation. As a result, the average MAPE of study cases was 9.05%, which was found to be 1.27 times more accurate on average than the nearest ASOS, respectively. In particular, it was confirmed that the inverse distance weighted interpolation (IDW) prediction error was the lowest in July, and the accuracy of the prediction was higher in regions with denser contours of predicted solar radiation. 분산 그리드에서 태양광발전시스템의 발전량을 예측하고 관리하기 위해서 일사량 예측이 필수적이다. 일사량 예측모델의 정확도를 높이려면 발전소가 위치한 대상 위치에서 측정된 데이터를 활용해야 하지만 관측 데이터가 없는 경우에는 정부에서 운영하는 종관기상관측소(ASOS), 방재기상관측소(AWS) 등의 공공데이터를 효과적으로 활용할 수 있다. 하지만 목표 위치가 관측소에러 멀리 떨어져 있는 경우 거리의 차이로 인해 예측의 불확실성이 커질 것으로 예상된다. 본 연구에서는 이러한 문제를 해결하기 위하여 샘플링되지 않는 위치의 값을 추정할 수 있는 공간 통계기법인 역거리 가중 보간(IDW)를 활용하여 일사량을 추정하였다. 또한 실제 운영중인 6개 태양광발전소를 사례연구로 예측모형을 검증하여 역거리 가중 보간법(IDW)의 국내 적용 가능성을 확인하였다. 그 결과, 6개 연구 대상의 평균 MAPE는 9.05%였으며, 이는 각각 최근접한 ASOS 대비 평균 1.27배 더 정확한 것으로 나타났다. 특히, 7월에 역거리 가중 보간(IDW) 예측 오차가 가장 낮았으며, 예측된 일사량의 등고가 조밀한 지역일수록 예측의 정확도가 높은 것으로 확인되었다.

공간통계기법을 이용한 전국 일 최고/최저기온 공간변이의 추정

신만용 ( Man Yong Shin ),윤진일 ( Jin Il Yun ),서애숙 ( Ae Sook Suh ) 大韓遠隔探査學會 1999 大韓遠隔探査學會誌 Vol.15 No.1

농업을 비롯한 산업활동을 효율적으로 수행하기 위해서는 전문 기상정보의 활용이 필수적이다. 영농활동에 있어서 의사지원시스템의 핵심으로 떠오르고 있는 작물 생장모형은 부단히 변화하는 대기환경에 대한 공간정보를 요구하기 때문에, 모형의 실용화를 위해서는 기상 관측밀도가 낮은 광범위한 작물 생육지역을 대상으로 일별 기상요소에 대한 공간분포를 추정해야 한다. 이러한 취지에서 본 연구는 미관측 지점을 포함하는 우리 나라 전국을 대상으로 작물모형의 구동에 필요한 최소 기상요소들 중에서 일 최고 및 일 최저기온의 공간적인 분포를 추정화고 그 추정 精度를 검증하고자 하였다. 이를 위해 먼저 58개 지점의 23년간 실측 기온자료로부터 지형기후학적 방법에 의하여 격자단위의 월별 기온평년값을 추정하고, 조화해석법에 의하여 일별값으로 변환하였다. 66개 기상청관측소에서 수집된 임의 날짜의 최고/최저기온값과 관측소 해당 격자점의 평년값간 편차를 구한 다음, 미관측 격자점을 포함하는 한반도 전역의 기온 편차를 거리역산기중법에 의하여 내삽·추정하였다. 각 격자점의 최종적인 기온 추정값은 기온 평년값에 이 편차를 더함으로써 얻었다. 얻어진 온도분포는 위성자료로부터 추정한 지표온도분포 양상과 크게 다르지 않았다. 300 여개의 자동기상관측장비들로부터 수집된 자료와 비교한 결과, 추정오차는 1.5℃~2.5℃였다. The use of climatic information is essential in the industrial society. More specialized weather services are required to perform better industrial activities including agriculture. Especially, crop models require daily weather data of crop growing area or cropping zones, where routine weather observations are rare. Estimates of the spatial distribution of daily climates might complement the low density of standard weather observation stations. This study was conducted to estimate the spatial distribution of daily minimum and maximum temperatures in Korean Peninsula. A topoclimatological technique was first applied to produce reasonable estimates of monthly climatic normals based on 1km × 1km grid cell over study area. Harmonic analysis method was then adopted to convert the monthly climatic normals into daily climatic normals. The daily temperatures for each grid cell were derived from a spatial interpolation procedure based on inverse-distance weighting of the observed deviation from the climatic normals at the nearest 4 standard weather stations. Data collected from more than 300 automatic weather systems were then used to validate the final estimates on several dates in 1997. Final step to confirm accuracy of the estimated temperature fields was comparing the distribution pattern with the brightness temperature fields derived from NOAA/AVHRR. Results show that differences between the estimated and the observed temperatures at 20 randomly selected automatic weather systems(AWS) range from -3.0℃ to +2.5℃ in daily maximum, and from -1.8℃ to +2.2℃ in daily minimum temperature. The estimation errors, RMSE, calculated from the data collected at about 300 AWS range from 1.5℃ to 2.5℃ for daily maximum/minimum temperatures.

지역별 잣나무 초기생장에 미치는 미기수의 영향 : 연년생장과 미기후와의 관계 Relation between Annual Increment and Local Climatic Conditions

전상근,신만용 한국농림기상학회 1999 한국농림기상학회지 Vol.1 No.1

This study was conducted to investigate the effects of local climatic conditions on the annual increment of Korean white pine planted in Gapyung and Young dong. For this, stand variables such as mean DBH, mean height, basal area per hectare, and volume per hectare by stand age were measured and summarized for each locality. Based on these statistics, annual increments for 8 year s from stand age 10 to 18 were calculated for each of stand variables. A topoclimatological technique which makes use of empirical relation ships between the topography and the weather in study sites was applied to produce normal estimates of monthly mean, maximum, minimum temperatures, relative humidity, precipitation, and hours of sunshine. Then, the yearly climatic variables from 1990 to 1997 for each study site were derived from the spatial interpolation procedures based on inverse-distance weighting of the observed deviation from the climatic normals at the nearest 11 standard weather stations. From these estimates, 17 weather variables such as warmth index, coldness index, index of aridity etc., which affect the tree growth, were computed on yearly base for each locality. The deviations of measured annual increments from the expected annual increments for 8 years based on yield table of Korean white pine were then correlated with and regressed on the yearly weather variables to examine effects of local climatic conditions on the growth. Gapyung area provides better conditions for the growth of Korean white pine in the early stage than Young dong area. This indicates that the conditions such as low temperature, high relative humidity, and large amount of precipitation provide favor environment for the early growth of Korean white pine. According to the correlation and regression analysis using local climatic conditions and annual increments, the growth pattern of Gapyung area corresponds to this tendency. However, it was found that the relationship between annual increments and local climatic conditions in Young dong area shows different tendency from Gapyung. These results mean that the yearly growth pattern could not sufficiently be explained by climatic conditions with high variance in yearly weather variables. In addition, the poor growth in Young dong area might not only be affected by climatic conditions, but also by other environmental factors such as site quality.

지역별 잣나무 초기생장에 미치는 미기후의 영향 - 연년생장과 미기후와의 관계-

전상근,신만용,Chon Sang- Keun,Shin Man Yong 한국농림기상학회 1999 한국농림기상학회지 Vol.1 No.1

본 연구는 경기도 가평과 충천북도 영동에 식재된 잣나무 임분의 임령별 연년생장을 파악하고 지역적 기후조건이 잣나무 초기생장에 미치는 영향을 구명하기 위하여 수행하였다. 먼저 각 지역의 임령별 평균 흉고직경, 평균 수고, ha당 흉고단면적, ha당 재적 등의 임분통계량을 측정 요약하였으며, 이에 근거하여 10년생부터 18년생까지 8년간의 초기생장에 대한 임분변수별 연년생장량을 파악하였다. 연구 대상지의 지역별 미기후는 지형과 기후치간의 관계를 이용한 지형기후학적 방법에 의해 평균기온, 최고기온, 최저 기온, 상대습도, 강수량, 그리고 일조시수의 월별 평년 값을 추정한 후, 공간통계기법을 적용하여 지역별·연도별 월별 기후치를 추정하였다. 이들 자료를 이용하여 임목생장에 영향을 미칠 것으로 판단되는 온량지수, 한랭지수, 건조지수 등의 17개 기후변수를 지역별·연도별로 산출하고 임분변수별 연년생장량과의 상관분석 및 회귀분석을 실시하였다. 잣나무의 초기생장은 경기도 가평이 충청북도 영동에 비하여 훨씬 우수한 성장을 보이고 있었다. 일반적으로 잣나무 임분의 생장은 기온이 낮고 강수량이 많아 높은 습도를 유지하는 지역이 적합한 것으로 알려져 있다. 지역별의 추정된 연도별 미기후와 연년생장과의 상관관계와 회귀분석 결과에 의하면, 가평 지역 잣나무 유령임분의 연년생장은 이러한 일반적인 생장-기후 관계와 일치하는 것으로 나타났다. 하지만 영동지역의 임분변수별 연년생장과 미기후와의 관계는 가평지역과는 다소 다른 경향을 보이고 있다. 이러한 결과는 연간 변이가 심한 기후조건에도 불구하고 비교적 짧은 기간 동안의 자료만으로는 연년생장에 미치는 영향을 구명하는데 한계가 있다는 사실에 기인한다. 또한 영동지역의 저조한 생장의 원인에 미기후 조건 이외에도 지위와 같은 다른 환경요인이 복합적으로 작용하기 때문으로 판단된다. This study was conducted to investigate the effects of local climatic conditions on the annual increment of Korean white pine planted in Gapyung and Yaungdong. For this, stand variables such as mean DBH, mean height, basal area per hectare, and volume per hectare by stand age were measured and summarized for each locality. Based on these statistics, annual increments for 8 years from stand age 10 to 18 were calculated for each of stand variables. A topoclimatological technique which makes use of empirical relationships between the topography and the weather in study sites was applied to produce normal estimates of monthly mean, maximum, minimum temperatures, relative humidity, precipitation, and hours of sunshine. Then, the yearly climatic variables from 1990 to 1997 for each study site were derived from the spatial interpolation procedures based on inverse- distance weighting of the observed deviation from the climatic normals at the nearest 11 standard weather stations. From these estimates, 17 weather variables such as warmth index, coldness index, index of aridity etc., which affect the tree growth, were computed on yearly base for each locality. The deviations of measured annual increments from the expected annual increments for 8 years based on yield table of Korean white pine were then correlated with and regressed on the yearly weather variables to examine effects of local climatic conditions on the growth. Gapyung area provides better conditions for the growth of Korean white pine in the early stage than Youngdong area. This indicates that the conditions such as low temperature, high relative humidity, and large amount of precipitation provide favor environment for the early growth of Korean white pine. A ccording to the correlation and regression an analysis using local climatic conditions and annual increments, the growth pattern of Gapyung area corresponds to this tendency. However, it was found that the relationship between annual increments and local climatic conditions in Youngdong area shows different tendency from Gapyung. These results mean that the yearly growth pattern could not sufficiently be explained by climatic conditions with high variance in yearly weather variables. In addition, the poor growth in Youngdong area might not only be affected by climatic conditions, but also by other environmental factors such as site quality.

공공 빅데이터를 활용한 행정 서비스 개선 연구 : 유성구 여성안심귀갓길 효과 분석

나웅재,박여진,양서연,김태구 한국서비스경영학회 2024 서비스경영학회지 Vol.25 No.1

여성안심귀갓길은 여성을 대상으로 한 범죄를 예방하기 위해 도입된 대표적인 범죄예방환경설계(CPTED)정책이다. 본 연구는 공공데이터를 활용하여 대전 유성구에 설치된 여성안심귀갓길과 여성안심시설물의 범죄 예방 효과를 분석하였다. 유성구를 100m 단위의 그리드로 나누어 대중교통 이용량, 유흥업소 수 등을 CPTED 설치 여부와 함께 사용하여 실제 범죄 신고 건수에 대한 영향력을 검증하였다. 또한 각 변수의 그리드별 자료를 역거리 가중 보간법을 통해 인근 그리드로 확산 적용하여 현실적으로 지리정보를 반영하였다. 분석 결과, 여성안심귀갓길과 여성안심시설물을 모두 설치할 경우 가장 범죄 예방 효과가 큰 것으로 나타났다. 최종적으로 본 연구에서는 분석 결과를 활용하여 새로운 CPTED 시설의 도입을 위한 최적의 후보지를 제안하였다. The Women's Safe Way Home is a representative CPTED policy to prevent crimes against women. This study analyzed the crime prevention effects of the women's safe way home and women's safety facilities installed in Yuseong-gu, Daejeon, using public data. The entire Yuseong-gu was divided into grids of 100 meter increments, and public transportation usage and the number of accommodations and restaurants were used along with CPTED installation status to statistically verify their impacts on the actual number of crime reports. In addition, the grid-specific data for each variable were spread and applied to nearby grids through inverse distance-weighted interpolation to reflect geographical information realistically. The analysis results showed that the crime prevention effect was greatest when both women's safe way home and women's safety facilities were installed. Finally, the analysis results were used to suggest the best candidate locations for the introduction of new CPTED facilities.

미계측 유역의 유량지속곡선 산정을 위한 확장역거리가중법의 적용성 평가

유지수,무하마드 와심,신지예,김태웅 한국방재학회 2015 한국방재학회논문집 Vol.15 No.3

This study evaluated the availability of a geographical and physiographical space-based extended inverse distance weighting (EIDW) method for constructing a flow duration curve (FDC) at ungauged basin. The method relies on two issues; (1) selecting an appropriate model for the approximation of FDCs at homogeneous gauging sites and (2) presenting significant interpolations to predict the streamflow quantiles at ungauged site. This study investigated the performance of three parsimonious parametric models such as the linear, exponential and logarithmic model. This study then introduced the extended inverse distance weighting to develop a reliable ensemble averaging (REA) model through adapting three parametric models. The extended inverse distance weighting enables us to obtain optimal interpolation and high prediction efficiency after considering various range of distance decay parameter. Based on the cross validation using the root mean square error (RMSE) and Nash-Sutcliffe efficiency (NSE), the overall results revealed that the model with extended inverse distance weighting can provide more accurate, consistent and unbiased estimates of hydrologic quantiles compared to the traditional IDW. 본 연구는 미계측유역의 유량지속곡선의 산정을 위하여 지리·지형적 공간을 기반으로 하는 확장역거리가중법 (EIDW)의 적용성을 평가하였다. 본 연구는 (1) 균일한 계측 지역에서의 유량지속곡선을 작성하는 적절한 모형을 선택하고 (2) 미계측 지점에서 유량 분위값(quantile)을 추정하는 보간법을 제시하는 두 가지 이슈에 초점을 두고 있다. 이를 위해서 본 연구는 유량지속곡선을 작성하는 세 가지모형, 즉 선형모형, 지수모형, 로그모형의 적용성을 검토하고, 이를 바탕으로 신뢰성 있는 앙상블 평균 모형을 개발하기 위해서 확장역거리가중법을 적용하였다. 확장역거리가중법은 다양한 범위의 거리 감쇠계수를 고려할 수 있어 최적의 보간과 높은 예측효율을 가질수 있게 하였다. RMSE와 NSE를 이용한 교차검증 결과, EIDW 모형은 기존의 IDW 방법보다 더 정확하고 일관성 있으며 편향되지 않은 결과를 가지는 것으로 나타났다.

실시간 도로기상 위험지수의 시각화 방법

조재경,지가현,윤상후 한국자료분석학회 2018 Journal of the Korean Data Analysis Society Vol.20 No.5

The weather on the highway affects the safety of the driver. Visibility is a distance for recognizing objects, which affects road safety. It is affected by rain, fog, snow, and particle matters. This study is dealt with a method for visualization of the road weather risk index that considering the visibility distance collected from weather stations near the highway and the driving speed. The stopping distance was calculated which considered the speed of the highway, the friction of the tire, and the reaction time of the driver. Because the visibility of the highway is not directly observed, it was reproduced by the inverse distance weighting method. As a result of evaluating the prediction performance, RMSE (root mean square error) was 1.5 or less using the inverse distance weighting method between 2015 and 2016. The ratio of visibility and stopping distance was visualized by plotgooglemaps. We hope that the real-time road weather risk index, which takes into account the visibility and the speed of a car, warns drivers to reduce the risk of safety accidents. 고속도로의 기상은 운전자의 안전에 영향을 미친다. 시정거리는 사물을 인식하기 위한 거리로 고속도로의 시정거리가 짧으면 대형사고가 발생할 가능성이 높다. 시정거리는 비, 안개, 눈, 미세먼지 등에 영향을 받는다. 본 연구는 고속도로 인근의 종관기상관측장비에서 수집된 시정거리와 한국도로공사에서 제공되는 실시간 주행속도를 고려한 도로기상 위험지수를 생산하여 이를 시각화하였다. 연구를 위해 고속도로의 주행속도, 타이어의 마찰력, 그리고 운전자의 반응시간이 고려된 정지시거가 이용되었다. 고속도로의 시정거리는 직접 관측되지 않으므로 역거리가중법을 통해 고속도로 단위로 재생산되었다. 2015년부터 2016년까지 2년간 역거리가중법으로 고속도로의 시정거리 예측성능을 평가한 결과 RMSE가 1.5 이하였다. 고속도로 단위의 시정거리는 주행속도가 고려된 정지시거의 비를 통해 도로기상 위험지수로 재생산되어 Google 지도로 시각화되었다. 실시간 시정거리와 주행속도가 고려된 도로기상 위험지수는 고속도로 운전자들에게 시정악화에 따른 대형 교통사고의 위험성을 경고하므로 안전사고의 피해를 감소시키는 데 도움이 된다.

인천 송도지역 지층분포 추정을 위한 크리깅과 역거리가중치법의 적용 한국지반공학회논문집 제26권 제1호인천 송도지역 지층분포 추정을 위한 크리깅과 역거리가중치법의 적용

김동휘,류동우,최영민,이우진 한국지반공학회 2010 한국지반공학회논문집 Vol.26 No.1

Geo-layer information is important to determine pile length and estimate residual settlement in the construction site. An overall spatial distribution of geo-layers in the entire construction site can be predicted using drill-log information. In this study, the geo-layer distribution at Song-do area was estimated by kriging and inverse distance weighting methods, and a cross validation was adopted to verify the reliability of estimation results. The analysis results indicate that the best fitted theoretical variogram model to the experimental variogram does not always provide the most reliable estimation in the kriging method. The proper value of inverse distance weighting method must be determined by types of geo-layer, because the value is affected by types of geo-layer. Results of the kriging method show more reliable results than those of inverse distance weighting method, and the structure of geo-layer distribution could be evaluated by variogram in the kriging method.

Development of a Virtual Reference Station-based Correction Generation Technique Using Enhanced Inverse Distance Weighting

태현우,김혜인,박관동 사단법인 항법시스템학회 2015 Journal of Positioning, Navigation, and Timing Vol.4 No.2

Existing Differential GPS (DGPS) pseudorange correction (PRC) generation techniques based on a virtual reference station cannot effectively assign a weighting factor if the baseline distance between a user and a reference station is not long enough. In this study, a virtual reference station DGPS PRC generation technique was developed based on an enhanced inverse distance weighting method using an exponential function that can maximize a small baseline distance difference due to the dense arrangement of DGPS reference stations in South Korea, and its positioning performance was validated. For the performance verification, the performance of the model developed in this study (EIDW) was compared with those of typical inverse distance weighting (IDW), first- and second-order multiple linear regression analyses (Planar 1 and 2), the model of Abousalem (1996) (Ab_EXP), and the model of Kim (2013) (Kim_EXP). The model developed in the present study had a horizontal accuracy of 53 cm, and the positioning based on the second-order multiple linear regression analysis that showed the highest positioning accuracy among the existing models had a horizontal accuracy of 51 cm, indicating that they have similar levels of performance. Also, when positioning was performed using five reference stations, the horizontal accuracy of the developed model improved by 8 ~ 42% compared to those of the existing models. In particular, the bias was improved by up to 27 cm.

Development of a Virtual Reference Station-based Correction Generation Technique Using Enhanced Inverse Distance Weighting

Tae, Hyunu,Kim, Hye-In,Park, Kwan-Dong The Institute of Positioning 2015 Journal of Positioning, Navigation, and Timing Vol.4 No.2

Existing Differential GPS (DGPS) pseudorange correction (PRC) generation techniques based on a virtual reference station cannot effectively assign a weighting factor if the baseline distance between a user and a reference station is not long enough. In this study, a virtual reference station DGPS PRC generation technique was developed based on an enhanced inverse distance weighting method using an exponential function that can maximize a small baseline distance difference due to the dense arrangement of DGPS reference stations in South Korea, and its positioning performance was validated. For the performance verification, the performance of the model developed in this study (EIDW) was compared with those of typical inverse distance weighting (IDW), first- and second-order multiple linear regression analyses (Planar 1 and 2), the model of <xref>Abousalem (1996)</xref> (Ab_EXP), and the model of <xref>Kim (2013)</xref> (Kim_EXP). The model developed in the present study had a horizontal accuracy of 53 cm, and the positioning based on the second-order multiple linear regression analysis that showed the highest positioning accuracy among the existing models had a horizontal accuracy of 51 cm, indicating that they have similar levels of performance. Also, when positioning was performed using five reference stations, the horizontal accuracy of the developed model improved by 8 ~ 42% compared to those of the existing models. In particular, the bias was improved by up to 27 cm.